In order to clarify a guideline for designing of composition of a flux which can achieve both minimizing a refractory corrosion by the flux and maximizing of solubility of Mn oxides into the flux, corrosion tests for refractory was conducted in the atmosphere. The basic composition of flux is Na₂O-B₂O₃ and the refractory is Mullite (3Al₂O₃・2SiO₂), assuming a process of melting of a copper alloy containing Mn as easily oxidized elements. Although the corrosion ratio of refractory became larger with increasing of mole fraction of Na₂O in flux, the concentration of refractory's constituents in the flux have different tendency predicted by the results of corrosion ratio. Through the corrosion test, the Na₂O-B₂O₃ based flux has penetrated inside the refractory with Mn, and a part of that Mn has reacted with Al₂O₃ to form MnAl₂O₄. However, in the refractory/flux interface no clear formation of the compound layer could be confirmed due to the reaction between the refractory's constituents and the flux. In addition, the relationship between the corrosion ratio and the equilibrium solubility of 3Al₂O₃・2SiO₂ for Na₂O-B₂O₃ flux calculated by thermodynamic database was investigated. The result shows that there is not a clear relationship between them. The cause of this can be explained by the affection of corrosion inside the refractory by the penetration of the flux through the pores in the refractory. Furthermore, it was shown that the amount of Mn oxide dissolved in the flux was strongly affected by the viscosity of the flux by calculation.

Consequently, in order to design a proper composition of flux in this study, it became clear that the thermodynamic approach alone was not enough and more detailed examinations such as the wettability between the flux and refractory, properties of flux, especially penetration phenomena were also important.

The evolution of microstructure during the isothermal aging at 1173 K was investigated for the wrought Ni-based superalloy Udimet 520 solution-treated at 1393 K for 4 h followed by various cooling rates. The age-hardening behavior was observed during the isothermal aging for the water-quenched (WQ) and air-cooled (AC) specimens after the solution treatment, while it could not be detected for the furnace-cooled (FC) specimen. No primary γ′ particles were observed in any continuously cooled samples. For the WQ and AC specimens, the size of the secondary γ′ precipitates increased during the isothermal aging along the Ostwald ripening and their morphology evolved from a spherical shape to an intermediate shape between spherical and cuboidal ones. On the contrary, the secondary γ′ particles exhibited an octodendritic shape for the as FC specimen, and the octodendritic character of the secondary γ′ particles was emphasized during the isothermal aging resulting in the splitting of the secondary γ′ particles. It was found that the splitting of γ′ particles occurs during the isothermal aging for the Alloy 80A with a lower volume fraction of γ′ phase around 20％.

To investigate the reason why low-carbon steels with carbon-clusters shows the maximum strength during low-temperature aging, interactions between an edge dislocation and carbon clusters are performed through molecular dynamics (MD) simulations. Carbon clusters are modeled based on atom probe tomography (APT) observations. To express a transition process of carbon configurations from solid solution state to carbon cluster state to precipitation state during aging process, we reduce a carbon presence area with a fixed number of carbon atoms, i.e., the carbon concentration can be continuously increased. The MD simulations can represent the age hardening/softening tendency observed in the experiment and the carbon cluster state shows the maximum strength where the dislocation passes through the carbon cluster not by the Orowan but by the cutting mechanism. The MD analysis found that partial clusters in the carbon cluster act as the main resistance to dislocation passage; the biased distribution of carbon atoms is also confirmed in the actual observed carbon clusters by APT. A new interaction mechanism between dislocation and carbon clusters is developed based on the phenomena in the MD simulations and the availability is discussed.

Fine particle peening (FPP) is a surface modification process in which fine hard metallic particles project onto the substrate. During this process, shot particles that remain transfer to the substrate and form a complex and multi-layered lamellar structure in which transferred fragments are distributed in the depth direction and several tens of micrometers from the surface. Because the bond strength at each interface of this structure and its generation mechanism are unclear, the local bond strength of the laminated interface was evaluated in this study by a microscale tensile test conducted by focus ion beam facilities. In the transferred lamellar modified region, the hardness was considerably higher than that of as-received materials, and the microscopic bonding strength between the transferred copper or iron and the aluminum matrix was 300 MPa or greater. When steel particles were used with FPP, iron oxide was mainly transferred to the aluminum matrix. However, copper plating on the particles suppressed the oxidation of the steel composing the particles and, as a result, the metallic steel was transferred. The bonding strength was improved by oxidation-controlled steel particles.